---
title: "Airport Dashboard"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    vertical_layout: fill
    theme: spacelab
    social: menu
    source_code: embed
---

```{r setup, include=FALSE}

list.of.packages <- c("ggplot2", "flexdashboard", "ggmap", "maps", "mapdata", "reshape2", "stringr", "scales", "plotly", "plyr")
new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])]
if(length(new.packages)) install.packages(new.packages, dependencies = TRUE)
library(flexdashboard)
library(ggplot2)
library(ggmap)
library(maps)
library(mapdata)
library(reshape2)
library(stringr)
library(scales)
library(plotly)
library(plyr)
```

Volume
======================================================================= 

Sidebar {.sidebar}
-----------------------------------------------------------------------

**Dashboard Navigation:**

Use the **Volume** tab to explore airports with the busiest departures and arrivals by volume

Use the **Delays by Origin** tab to explore some of the worst offenders in delayed flights, most freuqent cause of delays, and the worst day of the week to fly in popular airports

Use the **Delays by Route** tab to explore which routes experience the most delays 

Use the **Weather Metrics** tab to explore how precipitation in large cities affect filghts in those cities 

Use the **Forecasted Delays** tab to explore how likely a plane is likely to be delayed due to precipitation 



**Notes:**

Any and all graphs can be filtered by clicking on the airport or city in the margin

For optimal view of the dashboard, please maximize your window

Data is based on historical airport and weather data from the last 10 years 




**Authors:** 

Michael Alexander, Mike Amodeo, Roseanna Hopper, Yifan Sun



Column {.tabset}
-----------------------------------------------------------------------

### Top 25 US Airports by Departure Volume

```{r}
airports <- read.csv("https://raw.githubusercontent.com/jpatokal/openflights/master/data/airports.dat", header = FALSE)
colnames(airports) <- c("ID", "name", "city", "country", "IATA_FAA", "ICAO", "lat", "long", "altitude", "timezone", "DST")
busydep <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/top25airportsdep.csv", header = FALSE)
colnames(busydep) <- c("ID", "Origin", "X2007", "X2008", "X2009", "X2010", "X2011", "X2012", "X2013", "X2014", "X2015", "X2016")
busyarr <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/top25airportsarr.csv", header = FALSE)
colnames(busyarr) <- c("ID", "Dest", "X2007", "X2008", "X2009", "X2010", "X2011", "X2012", "X2013", "X2014", "X2015", "X2016")
busydep <- busydep[-1,]
busyarr <- busyarr[-1,]
years <- c(colnames(busydep)[3: ncol(busydep)])  
usairports <- subset(airports, country == "United States")
busydep[, years] <- lapply(busydep[, years], function(x) as.numeric(as.character(x)))
busyarr[, years] <- lapply(busyarr[, years], function(x) as.numeric(as.character(x)))
busydepairports <- merge(busydep, usairports, by.x=c("Origin"), by.y=c("IATA_FAA"))
busydepairports$meanvol <- rowMeans(busydep[, years], na.rm = T)


usmap <- get_map(location = "United States", zoom=4, maptype="terrain",
                 source = "google", color = "color")

airportmapdep <- ggmap(usmap) + geom_point(aes(x = long, y=lat, size=meanvol, color = Origin), data=busydepairports, alpha = 1.0) +
  theme(legend.position = "none", axis.title = element_blank(), axis.text = element_blank(), axis.line = element_blank())

airportmapdep

```


### Top 25 US Airports by Arrival Volume
```{r}

busyarrairports <- merge(busyarr, usairports, by.x=c("Dest"), by.y=c("IATA_FAA"))

years = c(colnames(busyarr)[3: ncol(busyarr)])
busyarrairports$meanvol <- rowMeans(busyarr[, years], na.rm = T)


usmap <- get_map(location = "United States", zoom=4, maptype="terrain",
                 source = "google", color = "color")

airportmaparr <- ggmap(usmap) + geom_point(aes(x = long, y=lat, size=meanvol, color = Dest), data=busyarrairports, alpha = 1.0) +
  theme(legend.position = "none", axis.title = element_blank(), axis.text = element_blank(), axis.line = element_blank())

airportmaparr
```


Column {data-width=500}
-----------------------------------------------------------------------

### Average Annual Departure Volume

```{r}

dep <- busydep[, 2:ncol(busydep)]

deptimeseries1 <- melt(dep, id.vars="Origin", value.name = "value", variable.name = "Year")
deptimeseries1$Year<- substr(deptimeseries1$Year, 2,5) 

deptimeseries <- ggplot(data=deptimeseries1, aes(x=Year, y=value, group = Origin, color = Origin)) + geom_line() + geom_point(size=1, shape = 21, fill = "white") + labs(x="Year", y= "Annual Departure Volume") + scale_y_continuous(labels = comma) + theme(axis.text.x=element_text(angle=90, hjust=1))
deptimeseries

ggplotly(deptimeseries)

```

### Average Annual Arrival Volume

```{r}

arr <- busyarr[, 2:ncol(busyarr)]

arrtimeseries1 <- melt(arr, id.vars="Dest", value.name = "value", variable.name = "Year")
arrtimeseries1$Year<- substr(arrtimeseries1$Year, 2,5) 

arrtimeseries <- ggplot(data=arrtimeseries1, aes(x=Year, y=value, group = Dest, color = Dest)) + geom_line() + geom_point(size=1, shape = 21, fill = "white") + labs(x="Year", y= "Annual Arrival Volume") + scale_y_continuous(labels = comma) + theme(axis.text.x=element_text(angle=90, hjust=1))
arrtimeseries

ggplotly(arrtimeseries)

```





Delays by Origin
=======================================================================

Column {.tabset}
-----------------------------------------------------------------------

### Airports with Highest Proportion of Delayed Flights

```{r}
high_delay_prop <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/highest_delay_prop.csv", header = TRUE)

high_delay_prop <- plyr::arrange(high_delay_prop, origin)

highprop <- ggplot(data=high_delay_prop, aes(x=origin, y=delayprop, color=origin, fill=origin)) + geom_bar(stat = "identity") + 
  ggtitle("From Top 100 Airports by Departure Volume") + labs(x="Origin", y="Proportion of Delayed Flights") + scale_y_continuous(labels=comma)  + theme(axis.text.x=element_text(angle=90, hjust=1), plot.title= element_text(color="#666666", size=12, hjust=.7))  + theme(legend.position="none")
highprop
ggplotly(highprop)
```

### Airports with Highest Proportion of Long Delays (>1 hr.)

```{r}
long_delay_prop <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/highest_long_delay_prop.csv", header = TRUE)

long_delay_prop <- plyr::arrange(long_delay_prop, origin)

longprop <- ggplot(data=long_delay_prop, aes(x=origin, y=delayprop, color=origin, fill=origin)) + geom_bar(stat = "identity") + ggtitle("From Top 100 Airports by Departure Volume") + labs(x="Origin", y="Proportion of Flights Delayed > 1 Hour") + scale_y_continuous(labels=comma)  + theme(axis.text.x=element_text(angle=90, hjust=1), plot.title= element_text(color="#666666", size=12, hjust=.7))  + theme(legend.position="none")
longprop
ggplotly(longprop)
```



Column {data-width=500}
-----------------------------------------------------------------------

### Most Frequent Cause of Delay Over Time

```{r}
delay_cause <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/delay_cause_total_year.csv", header = TRUE)

delay_cause_timeseries <- melt(delay_cause, id.vars="delay_cause_total_year.year", value.name = "value", variable.name = "Cause")

levels(delay_cause_timeseries$Cause) <- c('Carrier', 'Weather', 'National Airspace System', 'Security', 'Late Aircraft')

delay_plot <- ggplot(data=delay_cause_timeseries, aes(x=delay_cause_total_year.year, y=value, group = Cause, color = Cause)) + 
  geom_line() + 
  geom_point(size=1, shape = 21, fill = "white") + 
  labs(x="Year", y= "Total Annual Delay Minutes") + 
  scale_color_discrete(name = "Cause of Delay") +
  scale_y_continuous(labels = comma) + 
  theme(axis.text.x=element_text(angle=90, hjust=1))

ggplotly(delay_plot)
```

### Worst Weekday Delays 

```{r}
weekday_delay <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/worst_dayofweek_final.csv", header = TRUE)

weekday_agg = as.data.frame(table(weekday_delay$worst_dayofweek_final.worst_dayofweek))

weekday_agg$Var1 <- factor(weekday_agg$Var1, levels= c("Sunday", "Monday", 
    "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday"))

weekday_plot <- ggplot(data=weekday_agg, aes(x=Var1,y=Freq,color=Var1,fill=Var1)) +
  geom_bar(position="dodge",stat="identity") + 
  theme(legend.position = "none") +
  labs(x="", y= "Airport Frequency") +
  coord_flip() +
  ggtitle("Worst Day for Delays") 

ggplotly(weekday_plot)
```


Delays by Route
=======================================================================

Column
-----------------------------------------------------------------------

### Routes with Longest Arrival Delays

```{r}
arrdelay_route <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/arr_delay_route.csv", header = TRUE)

arrdelay_route <- plyr::arrange(arrdelay_route, route)

arrdel_route <- ggplot(data=arrdelay_route, aes(x=route, y=avgarrdelay, color=route, fill=route)) + geom_bar(stat = "identity") + ggtitle("Longest Average Arrival Delays") + labs(x="Route", y="Arrival Delay (Minutes)") + scale_y_continuous(labels=comma)  + theme(axis.text.x=element_text(angle=90, hjust=1), plot.title= element_text(color="#666666", size=12, hjust=.5))+ theme(legend.position="none")
arrdel_route
ggplotly(arrdel_route)
```


### Routes with Longest Departure Delays
```{r}
depdelay_route <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/dep_delay_route.csv", header = TRUE)

depdelay_route <- plyr::arrange(depdelay_route, route)

depdel_route <- ggplot(data=depdelay_route, aes(x=route, y=avgdepdelay, color=route, fill=route)) + geom_bar(stat = "identity") + ggtitle("Longest Average Departure Delays") + labs(x="Route", y="Departure Delay (Minutes)") + scale_y_continuous(labels=comma)  + theme(axis.text.x=element_text(angle=90, hjust=1), plot.title= element_text(color="#666666", size=12, hjust=.5))+ theme(legend.position="none")
depdel_route
ggplotly(depdel_route)
```

Weather Metrics 
=======================================================================

Column {.tabset}
-----------------------------------------------------------------------

### Annual Days with Precipitation

```{r}
precip_days <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/Q1_precip_days_ann.csv", header = TRUE)
colnames(precip_days) <- c("Origin", "Annual_days")
precip_airports <- merge(precip_days, usairports, by.x=c("Origin"), by.y=c("IATA_FAA"))
airportmap_precip <- ggmap(usmap) + geom_point(aes(x = long, y=lat, size=Annual_days, color = Origin), data=precip_airports, alpha = 1.0) +
  theme(legend.position = "none", axis.title = element_blank(), axis.text = element_blank(), axis.line = element_blank())
airportmap_precip
```


Column {data-width=500}
-----------------------------------------------------------------------

### Percentage of Flights Delayed by Weather

```{r}
likelihood_monthly <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/Q2_likelihood_month.csv", header = TRUE)
colnames(likelihood_monthly) <- c("Origin", "January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December")
likelihood_melt <- melt(likelihood_monthly, id="Origin", value.name = "value", variable.name = "Month")
likelihood <- ggplot(data=likelihood_melt, aes(x=Month, y=value, group = Origin, color = Origin)) + geom_line() + geom_point(size=1, shape = 21, fill = "white") + labs(x="Month", y= "Percentage of Flights Delayed by Weather") + scale_y_continuous(labels = comma) + theme(axis.text.x=element_text(angle=90, hjust=1))
likelihood
ggplotly(likelihood)
```

### Likelihood of Delay due to Depth of Precipitation

```{r}
likelihood_depth <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/Q3_likelihood_based_on_depth.csv", header = TRUE)
colnames(likelihood_depth) <- c("Origin", "0 in", "1/4 in", "1/2 in", "1 in", "1.5 in", "2 in", "3 in", "4 in")
likelihood_depth_melt <- melt(likelihood_depth, id = "Origin", value.name = "likelihood", variable.name = "depth")
likelihood_D <- ggplot(data=likelihood_depth_melt, aes(x=depth, y=likelihood, group = Origin, color = Origin)) + geom_line() + geom_point(size=1, shape = 21, fill = "white") + labs(x="Depth of Rainfall", y= "Likelihood of Delay") + scale_y_continuous(labels = comma) + theme(axis.text.x=element_text(angle=90, hjust=1))
likelihood_D
ggplotly(likelihood_D)
```



Forecasted Delays 
=======================================================================
Column {data-width=500}
-----------------------------------------------------------------------


### 10 Day Delay Forecast Map

```{r}
forecast <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/Q4_forecast_delays.csv", header = TRUE)
#colnames(forecast) <- c("Origin", "Day 1", "Day 2", "Day 3", "Day 4", "Day 5", "Day 6", "Day 7", "Day 8", "Day 9", "Day 10")
forecast$Total <- rowSums(forecast[2:11])
forecast_delays <- melt(forecast, id = "Origin", value.name = "Delays", variable.name = "Day")
delays_airports <- merge(forecast, usairports, by.x=c("Origin"), by.y=c("IATA_FAA"))
forecast_map <- ggmap(usmap) + geom_point(aes(x = long, y=lat, size=Total, color = Origin), data=delays_airports, alpha = 1.0) +
  theme(legend.position = "none", axis.title = element_blank(), axis.text = element_blank(), axis.line = element_blank())
forecast_map
```

Column {data-width=500}
-----------------------------------------------------------------------

### 10 Day Forecasted Precipitation

```{r}
forecast_precip <- read.csv("https://raw.githubusercontent.com/r-hopper/W205-Final-Project/master/flight_app/final_csv/Q5_forecast_rain.csv", header = TRUE)
forecast_precip_melt <- melt(forecast_precip, id = "Origin", value.name = "Precipitation", variable.name = "Date")
forecast_graph <- ggplot(data=forecast_precip_melt, aes(x=Date, y=(Precipitation / 25.4), group = Origin, color = Origin)) + geom_line() + geom_point(size=1, shape = 21, fill = "white") + labs(x="Next 10 Days", y= "Precipitation (in)") + scale_y_continuous(labels = comma) + theme(axis.text.x=element_text(angle=90, hjust=1))
ggplotly(forecast_graph)
```


### Projected Delays in the Next 10 Days Due to Precipitation

```{r}
forecast_lines <- ggplot(data=forecast_delays[forecast_delays$Day != 'Total', ], aes(x=Day, y=Delays, group = Origin, color = Origin)) + geom_line() + geom_point(size=1, shape = 21, fill = "white") + labs(x="Next 10 Days", y= "Projected Number of Delays") + scale_y_continuous(labels = comma) + theme(axis.text.x=element_text(angle=90, hjust=1))
ggplotly(forecast_lines)
```